The cure horizon

The word "cure" is used carefully in medicine. A treatment that controls symptoms is not the same as a treatment that eliminates the underlying disease. Insulin manages type 1 diabetes; it does not cure it. Lifelong dietary restriction manages PKU; it does not cure it. Enzyme replacement therapy manages Pompe disease; it does not cure it. The patient's experience with all of these treatments is meaningfully better than the alternative trajectory. The biological condition that produces the symptoms persists.

A cure, in the strict sense, is the permanent resolution of the underlying biological cause. Onasemnogene abeparvovec, the AAV9-delivered gene therapy approved for SMA in 2019, is a cure if the SMN1 transgene continues to express across decades. The single dose replaces the missing gene, the motor neurons that were dying continue to function, and the disease does not progress. Whether the gene therapy meets the strict definition of cure depends on whether the transgene persists durably, which depends on data that does not yet extend across the timescale required to answer the question.

What the strict definition requires data to confirm

A cure trial measures whether the disease comes back. A management trial measures whether the symptoms stay controlled. The endpoints are different in kind, not just in magnitude. The cure trial requires longer follow-up because the durability question is the question. A 30-year follow-up of a child cured of SMA at age 1 is the kind of evidence that supports the cure claim. A 5-year follow-up answers a different and easier question.

The infrastructure to follow patients through cure and beyond is the infrastructure that does not currently exist for most rare diseases. Clinical trial registries follow patients during the trial. Disease-specific registries follow patients through clinical care. The post-cure cohort, where the patient is no longer symptomatically affected and may transition out of the specialty care infrastructure that captured their data, is invisible to most of the existing follow-up systems.

The patient cured at age 1 is healthy at age 5, has no metabolic clinic appointments at age 15, has long left pediatric care by age 25, and has perhaps a primary care physician who does not know the patient's history and would not capture it in any structured form even if it came up. The patient's data trajectory ends, in the registry sense, around the time the patient stops being symptomatic.

What follow-up across cure looks like

The infrastructure question is what kind of relationship the patient maintains with the data system after the cure. Three categories of post-cure data are required to support the durability assessment.

The first is the persistence assessment itself. Has the transgene continued to express? Has the protein continued to be produced? Has the biological function continued to be performed? The assessment requires laboratory measurements that the patient has to opt into receiving, processed at a laboratory equipped to make the measurements, with results that flow back to the data infrastructure. The frequency might be every few years for stable cases, or more often during the early years when persistence is less established.

The second is the late-effect monitoring. Gene therapies introduce vector DNA that integrates or persists episomally; the long-term consequences of that integration include theoretical risks that have to be characterized empirically. Insertional oncogenesis, immune response to the vector or the transgene product, off-target effects in tissues other than the intended target, and tissue-specific aging in the cured organ are all questions that require longitudinal monitoring.

The third is the comparison with unaffected cohorts. The cured patient's health trajectory should, if the cure is durable, resemble the trajectory of someone who never had the disease. Demonstrating that resemblance requires the cured patient's data to be comparable to data from age-matched unaffected populations. The comparability requirement constrains how the data is collected and what is measured.

What the data trust enables

The patient-controlled data trust supports post-cure follow-up in a way that condition-specific registries do not.

The relationship between the patient and the trust is durable. The trust holds the patient's data across changes in healthcare provider, insurance carrier, geographic location, and life circumstances. The patient who moves from pediatric to adult care does not move out of the data infrastructure. The patient who relocates internationally does not move out of the data infrastructure. The patient who is asymptomatic and has no current medical reason to interact with the rare disease care system retains the relationship with the trust.

The trust supports comparative analysis with unaffected cohorts. The same trust that holds rare disease patient data can hold control population data with appropriate consent. The cured patient's trajectory can be compared with the unaffected population's trajectory using the same data infrastructure, with statistical methods that account for the relevant confounders.

The trust persists across the development cycle of the next therapy. The cured patient's data, captured over decades, becomes the post-approval safety and efficacy evidence that supports the next regulatory action on the same therapy or on closely related therapies. The data that begins as confirmatory follow-up for one cure becomes natural history data for the next.

What the cure horizon means for the project

The infrastructure that supports cure follow-up is the infrastructure that supports everything else, with the longest time horizon attached to it. The dataset that follows patients across diagnosis, treatment, cure, and decades after cure is the dataset that has every analytic property the rare disease research community needs.

The construction of that dataset is the project the rare disease community has not been positioned to undertake. The pharmaceutical sponsor who pays for the trial does not pay for fifty-year follow-up. The academic registry funded by NIH does not span the careers of multiple principal investigators. The condition-specific advocacy organization does not have the resources to build long-horizon infrastructure.

The patient-controlled trust, by contrast, is structurally suited to the long horizon because the beneficiary is the affected community rather than any institutional sponsor. The trust persists across sponsor transitions, funding cycles, and personnel changes. The data accumulates because the contributors continue contributing across decades. The long horizon is the horizon the trust is designed for.

The cure horizon is not a marketing term. It is a design specification. The infrastructure has to be built to support the durability assessment that the cure question requires. The infrastructure that is built to support the durability assessment is the infrastructure that supports everything else along the way. The long horizon is the through line that connects the immediate research questions to the eventual ones, and the trust is the structure that holds across the time required to answer them.